Hong Kong Journal of Emergency Medicine

(1) To assess the normal range of thickness of the epiglottis by means of ultrasound measurement. (2) To evaluate inter-observer agreement in measuring the thickness of the epiglottis of normal individuals by ultrasound. (3) To assess the association between biological factors and the thickness of the epiglottis.

Fifty adult volunteers working at a local accident and emergency department were recruited. The thickness of the epiglottis was measured by means of ultrasound examination, which was performed twice by two emergency physicians at different time. The study subjects' age, sex, height and body weight were recorded.

The mean thickness of the epiglottis was 0.236 cm and the standard deviation was 0.020. Male subjects had thicker epiglottis. Interobserver agreement of the two emergency physicians who performed the ultrasound scan was very good. Multiple regression models showed that sex and height were useful predictors of the thickness of the epiglottis.

Bedside ultrasound assessment of the epiglottis is an easy, rapid and reliable method to evaluate its thickness. Further studies are needed to evaluate the thickness of the epiglottis in patients with epiglottitis before it can be put into clinical use.

To find out the characteristics and presentations of imperforate hymen in the local population in Hong Kong and to assess if diagnosis of imperforate hymen made in the emergency department can reduce time to operation and length of hospital stay.

Retrospective study.

A regional public hospital in Hong Kong.

We retrospectively collected data of all patients with the diagnosis of imperforate hymen (ICD 9 coding 752.42) from the period of January 1999 to June 2009. Demographics of the patients, their presenting symptoms and signs, the diagnostic process, investigation findings and time of operation were recorded and analysed.

Fifteen cases of imperforate hymen were reported during that period. All were adolescent girls aged from 10 to 15 years. A total of 13 patients presented to the accident and emergency department (AED). Seven patients had the diagnosis made in the AED and 2 patients received bedside pelvic ultrasound. More than half of them (8 patients) presented with acute retention of urine. Other presentations included lower abdominal pain, constipation, lower abdominal mass, and protruding introital mass. Most cases diagnosed in the AED (5 out of 7) were admitted to the gynaecology ward while the others were admitted to the surgical or urology wards. Five out of the 7 cases diagnosed in the AED received operation within 24 hours; whereas only 1 out of the 6 cases with the diagnosis made after admission had operation within 24 hours. The difference was statistically significant (p=0.035, Fisher's exact test). The mean length of stay of the group diagnosed in the AED was 1.9 days while the mean length of stay for the group diagnosed after admission was 4.2 days.

Early diagnosis of imperforate hymen and haematometrocolpos in the AED for adolescent girls with primary amenorrhoea could have positive impact on proper admission to the gynaecology ward, prompt operation and shorter length of stay in hospital.

Carbon monoxide (CO) is a potent myocardial toxin. We investigated the association between acute CO intoxication and electrocardiographic QT interval (QTmax/QTmin), corrected QT interval (cQTmax/cQTmin), QT dispersion (QTd) and corrected QT dispersion (cQTd), P wave duration (Pmax/Pmin) and P wave dispersion (Pd), which were known as predictors of ventricular arrhythmias, atrial fibrillation and sudden death.

Electrocardiography with 12-leads and blood gas were taken from 65 patients with CO intoxication as well as 65 control patients with similar age and gender distribution at the admission time to the emergency department and at the 4th hour post-therapy. The carboxyhaemoglobin (COHb) levels and the relationships with electrocardiographic parameters were studied.

The mean COHb levels were 21.43±6.85% and 1.37±0.98% in intoxicated patients and in the control group respectively. QTmax, QTd, cQTmax, cQTd, Pmax, and Pd were found significantly higher in intoxicated patients when compared to the control group (373.98±40.35 vs. 355.98±32.88, p=0.006; 49.29±22.66 vs. 20.43±11.16, p<0.001; 455.38±30.72 vs. 419.57±22.27, p<0.001; 60.88±25.99 vs. 25.75±13.13, p<0.001; 107.91±13.28 vs. 96.65±12.65, p<0.001, 46.85±13.49 vs. 35.22±10.07, p<0.001 respectively). In Pearson correlation analysis, there were significant correlations between COHb level and QTmax, QTd, cQTmax, cQTd, Pmax and Pd (r=0.211, p=0.016; r=0.610, p<0.001; r=0.599, p<0.001; r=0.628, p<0.001; r=0.458, p<0.001; r=0.414, p<0.001 respectively).

A correlation between reversible increases in QTd, cQTmax, cQTd, Pmax and Pd in the electrocardiogram and COHb can be observed in acute CO intoxication patients.

Congestive heart failure is a common disease presentation to the emergency department (ED) and most of these patients require hospitalisation for treatment. The aim of this study was to assess the predictive value of B-type natriuretic peptide (BNP) level measured at the ED with respect to the length of hospital stay (LOS) in patients presenting with congestive heart failure (CHF). We also compared whether the BNP level could be a better predictor of LOS than clinical parameters and attempted to establish a cut-off point of BNP level which could differentiate those CHF patients with short hospital stay from the whole population of CHF patients.

Patients with clinical features compatible with congestive heart failure were recruited into the study. Point-of-care test for BNP level was performed. The LOS, 6-month readmission and mortality were retrieved from the computer Clinical Management System of the hospital. Pearson's correlation coefficient was used to test the association of BNP level and the LOS, 6-month readmission and mortality rate.

One hundred and twelve (112) patients were recruited into this study and 20 of them succumbed during hospitalisation. The data of the remaining 92 patients were analysed. The correlation coefficient of BNP level and LOS was 0.088. The correlation coefficient of BNP level and readmission within 6 months was 0.130 while the correlation coefficient of BNP level and 6-month mortality was 0.121. The correlation coefficient of other clinical features and LOS ranged from −0.098 to 0.247. The mean LOS of patients with BNP less than 400 pg/ml was 93.29 hours (95%CI=57.38 to 129.19). The mean LOS of patients with BNP greater than 400 pg/ml was 181.05 hours (95%CI=126.01 to 236.08). Patients with BNP level less than 400 pg/ml tended to have a shorter LOS than those with BNP level greater than 400 pg/ml (p=0.046).

CHF patients with BNP level measured in the emergency department greater than 400 pg/ml tend to have a longer LOS than patients with BNP level less than 400 pg/ml. BNP is not a good predictor for readmission and mortality of these patients.

To evaluate the ability of the Modified Acute Physiology and Chronic Health Evaluation (M-APACHE II) scoring system for the prediction of mortality in patients admitted to a tertiary emergency department in Iran.

During the study period, all patients aged >12 years who had been admitted to the emergency department of a tertiary hospital in Tehran, Iran were enrolled in the study. Traumatic and poisoned patients and those who died immediately after arriving at the emergency room were excluded. Using the M-APACHE II, risk of mortality was calculated for each patient. Finally, expected and observed mortalities were compared and the accuracy of M-APACHE II for prediction of mortality was determined using receiver operating characteristics (ROC) analysis.

During the study period, 389 cases including 236 males (60.7%) were enrolled into the study. The mean age of the patients was 60.6±19.4 years (range 14 to 98 years). 117 patients died (30%) while the M-APACHE II predicted 129 deaths. The greatest discrepancy between observed and expected deaths occurred at M-APACHE II scores ≥21. The constructed area under the ROC curve basng on predicted and observed death was 0.938 (95% confidence interval 0.915-0.961).

M-APACHE II is an accurate scoring system for predicting mortality in patients admitted to the emergency department. However, further studies are needed to confirm our findings.

The purpose of this study was to identify risk factors predicting mortality in multiple blunt trauma patients so as to prompt appropriate management during trauma resuscitation.

To assess risk factors potentially related to mortality in multiple blunt trauma patients, we reviewed the medical records of 1419 multiple blunt trauma patients who were admitted to the emergency department. The patients were divided into two groups; the survival group (n=1308) and the death group (n=111). Initial data collected on arrival in the emergency department were analyzed.

In the study, 67.4% (n=956) of 1419 patients were male, 32.6% (n=463) were female. The average age was 21.19±0.50 years (range 1-92). After controlling for the factors significantly related to outcome (all p<0.05), death due to multiple blunt trauma was more likely in patients who were of older age, who had major chest injury, who had intra-abdominal solid organ injury and who had low Glasgow Coma Scale (GCS) score and low Revised Trauma Score (RTS).

We conclude that older age, major chest injury, intra-abdominal solid organ injury, low GCS and low RTS were identified as possible risk factors for mortality in multiple blunt trauma patients.

To study the local epidemiology and clinical presentation after ingestion of oral mucosal irritating plants.

Cases presenting with immediate oral mucosal irritation after plant product ingestion from July 2005 to June 2009 were retrieved from the Hong Kong Poison Information Centre's clinical database. Their clinical features and management were reviewed.

Thirty cases were retrieved and analysed, including 14 males and 16 females. Poisoning occurred all year round with no seasonal predilection. Vomiting and dysphagia were the two most common symptoms other than immediate oral mucosal irritation. Half of the cases could be managed with symptomatic treatment and discharged after a few hours of observation. Six patients presented with angioedema and were managed initially with the use of intravenous steroid, antihistamine and with or without adrenaline. Urgent airway assessment with laryngoscope was performed in 3 patients. No patient required endotracheal intubation.

Oral mucosal irritating plant ingestion is not uncommon in Hong Kong. Its clinical feature can mimic angioedema on presentation. Immediate and localized oral mucosal symptoms after plant product ingestion are the key to diagnosis.